An arteriovenous malformation (AVM) is a congenital vascular abnormality in the brain with direct connections between arteries and veins such that blood bypasses brain tissue. The primary presentation of AVM is intracranial hemorrhage which occurs in as many as 40-70% of patients and may lead to permanent injury or death. As the gold standard for the detection and evaluation of vascular malformations, conventional digital subtraction angiography (DSA) is an invasive procedure bearing risks of neurological complications, as well as risks of ionizing radiation and iodinated contrast. While superior for th delineation of vascular anatomy, DSA is not capable of providing quantitative assessments of blood flow or degree of shunt in an AVM. Existing MR techniques are suboptimal for quantifying the hemodynamics of vascular malformation, and the complexity of the vascular architecture is often inadequately demonstrated. Arterial spin labeling (ASL) is a noninvasive MRI technique that utilizes magnetically labeled blood water as an endogenous tracer for perfusion measurements. Due to the direct shunt between arteries and veins in AVMs, the labeled blood spins behave as an intravascular contrast agent, and can be utilized for visualizing the dynamic blood flow through feeding arteries, nidus and draining veins of an AVM. Furthermore, hemodynamic parameters such as blood flow, blood volume and mean transit time can be quantified by adapting the standard tracer kinetic model. We have recently developed such an entirely noninvasive and quantitative 4-D time-resolved dMRA technique by combining ASL with a segmented cine multiphase TrueFISP sequence. The goal of the present proposal is to further develop, validate and evaluate the clinical utility of 4-D non-contrast dMRA in assessing both the vascular architecture and hemodynamics of AVMs. In Aim 1, further technical development and optimization of 4-D non-contrast dMRA will be performed, including implementation of multi-bolus pulsed and pseudo-continuous ASL (pCASL) with vessel selective labeling; Cartesian sampling with view sharing; dynamic golden angle radial acquisition with k-space weighted image contrast (KWIC); in conjunction with parallel imaging and potentially compressed sensing. In Aim 2, validation of methods for quantifying blood flow and degree of shunt through AVMs using 4- D dMRA will be carried out by comparison with phase-contrast (PC) MRI and pCASL perfusion MRI. Finally in Aim 3, the clinical utility of the proposed 4-D dMRA technique will be evaluated in AVM patients by comparison with the reference standard of DSA, time-of-flight (TOF) MRA and T2 weighted MRI. Furthermore, repeated scans will be performed to test whether 4-D dMRA is able to detect changes of blood flow and degree of shunt through AVMs pre and post treatments. The proposed 4-D dMRA is expected to provide alternative and complementary approaches for conventional DSA and MRA/MRI techniques in quantitative assessments of hemodynamics in AVMs. It will be useful not only for evaluation of AVMs, but also for other cerebrovascular disorders such as steno-occlusive diseases and cerebral aneurysms. !